Pulsed-Laser Switching in the Bistability Domain of a Cooperative

Publication Date (Web): June 30, 2015. Copyright ... Strong Cooperative Spin Crossover in 2D and 3D Fe–M Hofmann-Like Coordination Polymers Based on...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/JPCC

Pulsed-Laser Switching in the Bistability Domain of a Cooperative Spin Crossover Compound: A Critical Study through Calorimetry M. Castro,*,† O. Roubeau,† L. Piñeiro-López,‡ J. A. Real,‡ and J. A. Rodríguez-Velamazán†,§ †

Instituto de Ciencia de Materiales de Aragón (ICMA), CSIC and Universidad de Zaragoza, Zaragoza, Spain Instituto de Ciencia Molecular (ICMol), Universidad de Valencia, 46980 Paterna, Valencia, Spain § Institut Laue-Langevin, Grenoble 38042 Cedex 9, France ‡

S Supporting Information *

ABSTRACT: The photoswitching from the low spin (LS) to high spin (HS) state and the reverse process in the bistability domain of spin crossover (SCO) compounds is a promising function to be used in molecular electronic devices, and evidenced mainly through spectroscopy. The phenomenon, and in particular its mechanism, is however still under debate since some controversial experimental results have been reported. Here we present a calorimetric experimental study of the photoswitching of the [Fe(pyrazine)Pt(CN)4] SCO material by a nanosecond-pulsed green laser. Our results confirm that the single laser pulse of varying energies results in significant LS to HS transformations and show that calorimetry provides an accurate quantification of the overall conversion. Successive pulses allow increasing the conversion, achieving a maximum of 60% under our experimental conditions. The HS to LS transformation is on the other hand not induced at any laser fluences contrary to previous reports. The results are compared with those reported with Raman spectroscopy and critically discussed in terms of efficiency of the transformation and potential thermal effects.



single green laser nanosecond pulse.5 Neither larger HS fractions, through additional pulses, nor the reverse switching, attempted with various wavelengths and fluences, could however be obtained. In the SCO complex [FeL(CN)2]·H2O (L = N3O2 Schiff-base macrocyclic ligand) the photoexcitation from a HS:LS (1:1, α = 0.5) mixture state to α ≈ 0.88 of HS species was produced using a single-shot laser pulse and monitored through absorption spectra.6 This work evidenced the existence of threshold values both in wavelength and in laser fluence to trigger the photoexcitation. The mechanism in this specific system however likely involves a change in coordination number and is thus not a simple SCO process. Several papers7−9 on the [Fe(pz)Pt(CN)4] SCO compound (pz = pyrazine) used Raman spectroscopy to monitor the effect of the light excitation. In this case and for a polycrystalline powder sample, a bidirectional photoinduced process (LS to HS and HS to LS) was obtained using the same nanosecondpulsed green laser.7 Photoswitching from the LS state was probed at several temperatures within the hysteresis loop reaching a HS fraction of α = 0.2 for a single laser pulse and 0.4 after a second pulse was applied. Surprisingly, no further increase in HS fraction was detected with additional pulses, while a quite large laser fluence of 250 mJ/cm2 was used. In a later study9 a 100% conversion could be reached using a much lower laser fluence of at least 20 mJ/cm2, with a threshold value for partial conversion of around 5 mJ/cm2. On a polycrystalline

INTRODUCTION The control of the magnetic properties of molecular compounds by external stimuli, including temperature, light, pressure, and electric or magnetic fields, is an important scientific topic spurred on by envisaged uses in molecular spintronics as memory storage, transistors, and switching devices.1 Among these switchable magnetic materials, FeII spin crossover (SCO) compounds2,3 are one of the most interesting systems, due to the spin transition between a diamagnetic low spin state (LS, S = 0) and a paramagnetic high spin state (HS, S = 2), associated with changes in various physical properties (optical, electrical, and structural), induced by an external perturbation. The well-studied switching induced by light, so-called light-induced excited spin state trapping (LIESST),4 is however limited to low temperatures by the thermally activated relaxation of the metastable trapped HS state. Recently, a novel approach has been considered to obtain photoinduced reversible spin state switching at higher temperature; it contemplates irradiation within the bistability domain of SCO materials with a large thermal hysteresis loop, characteristic of the presence of a first-order transition. In such conditions, the light-trapped species are expected to be stable, as long as the system is maintained within the bistability conditions. A few promising reports demonstrating the validity of the approach have already been published. The first experimental result was obtained measuring the optical reflectivity of the [Fe(PM-BiA)2(NCS)2] compound (PM-BiA = N-2′-pyridylmethylene-4-aminobiphenyl), for which a large stationary photoinduced HS fraction (α = 0.6) was produced within the thermal hysteresis loop from the initial LS state by a © XXXX American Chemical Society

Received: June 19, 2015 Revised: June 30, 2015

A

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

where the abrupt conversion to the HS state happens. Eventually, the dynamics of the growing or contraction of laser-induced initial HS domains and the related accommodation strain are very likely to play a role.22 Altogether, and although there is no doubt a single-laser pulse in the bistability domain of cooperative SCO materials does induce LS to HS phase transformations, the true mechanisms for both LS to HS and HS to LS transformations clearly remain disputable. In this respect, it should be noted that all reports so far have been based solely on (local) spectroscopic methods. Bulk structural, magnetic, and thermal studies remain clearly necessary to understand, improve, and apply the observed phenomena. In this work, we report the first calorimetric study of the laser-induced switching within the bistability domain of a SCO compound. We focus on the spin crossover [Fe(pz)Pt(CN)4] compound with its spin crossover around room temperature and test both the LS to HS and HS to LS phase transformations, seeking confirmation of the pulsed-laser switching phenomena and additional insights toward their understanding.

powder sample, the reverse transformation could only be produced from a HS−LS mixture state (with an initial LS fraction of 0.3).7 On the other hand, a complete bidirectional photoexcitation was obtained on single crystals, again using the same nanosecond-pulsed green laser.8 In this case, the HS to LS phase transformation was achieved from a full HS state and using a low laser fluence (≈5 mJ/cm2) that had no effect in the case of the polycrystalline sample. The photoexcitation within the hysteresis loop can also be performed using a continuous laser irradiation as shown by Raman spectroscopy with a complete LS to HS conversion of the [Fe(H4L)2](ClO4)2·H2O· 2(CH 3 ) 2 CO SCO compound (H 4 L = 2,6-bis(5-(2hydroxyphenyl)pyrazol-3-yl)pyridine).10 Single-pulse photoswitching was also attempted in triazole-based 1D polymeric spin crossover compounds.11 In [FeII(Htrz)2(trz)](BF4) (Htrz =1-H-1,2,4-triazole) a bidirectional switching, i.e., LS to HS and HS to LS, was detected through Raman spectroscopy by pulsed green irradiation at the same wavelength but only when the excitation was done at temperatures close to the ascending (378 K) and descending (351 K) branches of its large thermal hysteresis loop.12 No effect could however be detected when the excitation was performed at temperatures in the center of the hysteresis loop. In all these experimental reports the sample heating due to the irradiation was considered to have no significant effect on the detected phenomenology, although some unexplained aspects such as the reversibility of the process using an identical wavelength for which drastically different absorption densities are detected in the HS and LS states or the necessity to be close to (or even within) the thermal phase transition for realizing the switching may originate in thermally related effects. These would mostly result from a local (surface) overheating, over a very short time scale, and thus are difficult to detect and include a thermally induced LS to HS transformation due to sufficient heating or a thermally induced HS to LS transformation resulting from the response of temperature control to local heating. The former has indeed more recently been proposed as the origin of the impossibility to perform the HS to LS phase transformation in [Fe(PMBiA)2(NCS)2], whatever the wavelength and fluence used,13 as well as the main source of laser-induced switching in the SCO compounds [Fe(NH2-trz)3](NO3)2·H2O and [Fe(NH2-trz)3](Br)2·3H2O.14−16 In these latter studies, the authors used timeresolved optical reflectivity data, thus accessing only the surface response, and considered huge local heating. Additionally, a large local temperature gradient will produce a heat transfer to the rest of the material, increasing its temperature until a common temperature is reached when thermal equilibrium is obtained with the surrounding. This has in fact been directly observed with Raman microscopy with microscale spatial resolution, showing a LS to HS conversion outside the irradiated region in [FeII(Htrz)2(trz)](BF4).17 Although performed on a noncooperative FeIII SCO compound, timeresolved single-crystal optical and X-ray diffraction studies actually revealed for the first time a multistep process including photo, elastic, and thermal-induced phenomena occurring at different time scales.18,19 Recently, the same research group has also detected these consecutive three steps on a cooperative FeIII SCO compound20,21 (excitation was performed on the LS state but outside of the thermal hysteresis loop), and they conclude that a strong response in this case is mainly due to the temperature increase caused by the laser excitation which allows the sample to surpass its spin transition temperature



EXPERIMENTAL METHODS Synthesis and Sample Description. A microcrystalline powder (PP) and monocrystalline (SC) (ca. 40−80 μm) samples of the dehydrated form of the spin crossover compound [Fe(pyrazine)(Pt(CN)4)] were obtained as previously described.23 Both samples show orange color at room temperature indicative of a HS state. An aged PP sample (APP) taken from the same batch but after one year stored in a plastic bag at room temperature has been also used in some calorimetric experiments. Experimental Calorimetric Setup. A differential scanning calorimeter (DSC) model Q1000 from TA Instruments equipped with a liquid nitrogen cooling system (LNCS) in order to reach low temperature has been used. The irradiation of the sample was performed keeping open the oven and using the photocalorimetric commercial accessories (PCAs) that allow illuminating the sample inside the oven while it is under the helium gas atmosphere. In order to remove traces of water molecules possibly absorbed from moist air, the sample was submitted to an initial thermal treatment at 140 °C during 30 min inside the DSC oven and under helium gas atmosphere before proceeding with the calorimetric measurements. Irradiation was done when the calorimeter was operating in the isothermal mode; that is, the instrument was set to maintain constant the selected temperature (Tirradiation). Thermograms (cooling and heating) were performed at a scan rate of 10 K/ min. The sample masses were 1.10, 0.87, and 0.97 mg for PP, APP, and SC samples, respectively. The sample was spread on the aluminum pan (6 mm of diameter) and covered by a thin fused silica plate in order to avoid the loss of material. A pulsed Nd3+:YAG laser (LOTIS TII LS-2134U) operating at 532 nm with a pulse width of 7−8 ns was used as the excitation source. The spot size is 6 mm allowing the irradiation of the complete aluminum pan and therefore the whole sample. This whole photocalorimetric system is shown in Figure 1. In order to estimate the laser fluence (mJ/cm2), graphite powder (mass around 0.1 mg) was used as a good absorbing sample so the whole pulse energy is converted into heat. It was also spread on the pan as a thin film and covered by the fused silica plate, and the DSC heat flow peak, required for keeping constant the temperature after the laser pulse, was determined. B

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

temperature. Polycrystalline powders and single crystals of 1 or partially rehydrated 1 have very similar SCO properties with a large and rather abrupt hysteresis centered at room temperature; only slight variations of the transition temperatures have been reported, with values of Tcooling ranging 283−291 K and 1/2 ranging 305−311 K, and with a hysteresis width of 24− Theating 1/2 28 K.7,8,23,28 This compound has been extensively studied, and a collection of interesting properties have been reported like the chemoswitching29 between HS and LS states and molecular rotation correlated with the change of the spin state.30 The thermograms, i.e., heat flow vs temperature scans, obtained at 10 K/min are represented in Figure 2 for the studied PP (Figure 2A), APP (Figure 2B), and SC (Figure 2C) samples of [Fe(pyrazine)(Pt(CN)4)], respectively. These

Figure 1. Experimental photocalorimetric setup.

Results are shown in Figure S1 included in the Supporting Information. Additional Magnetic, Spectroscopic, and Structural Characterization. Magnetic measurements on the SC sample were performed with an MPMS-XL SQUID magnetometer from Quantum Design in the 200−340 K temperature range and under 1 T applied field. The SC sample was warmed to 140 °C for 30 min in a desiccator, before performing magnetic measurements. Single-crystal X-ray diffraction was performed at 310 and 240 K on the SC sample using an Oxford Diffraction Xcalibur diffractometer with enhanced Mo Kα radiation (λ = 0.71071 Å) at the X-ray and Fluorescence Analysis Service of the Servicio General de Apoyo a la Investigación-SAI, Universidad de Zaragoza. Cell refinement, data reduction, and absorption corrections were performed with the CrysAlisPro suite.24 The structure was solved with SIR9725 and refined on F2 with the SHELXTL suite.26,27 The visible absorption spectra of a powder sample near room temperature have been obtained for both LS and HS states by diffuse reflectance and using a Jasco V-60 UV/vis/NIR spectrophotometer equipped with a PIV-756/PIN 757 horizontal sampling integrating sphere system. The corresponding results provided in the Supporting Information are in agreement with previously reported data and support the spin transition as detected through DSC.



RESULTS AND DISCUSSION The three-dimensional porous coordination network [Fe(pz)Pt(CN)4] was first reported in 2001 in its hydrated form23 that exhibits a hysteretic SCO with Tcooling = 220 K and Theating = 1/2 1/2 240 K (T1/2 being the temperature at which HS and LS states have equal populations). Removal of the lattice solvent water molecules by heating to 120−130 °C for 30 min results in dehydrated [Fe(pz)Pt(CN)4] (1), which is stable when kept in a dry atmosphere and only slowly reabsorbs water from moist air. Dehydration can be done on single crystals without loss of crystallinity and results in a shift of the hysteresis toward higher

Figure 2. Heating and cooling thermograms for: (A) microcrystalline powder (PP) sample, (B) aged microcrystalline powder (APP) sample and (C) monocrystalline (SC) sample. C

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Table 1. Characteristic Temperatures and Thermodynamic Contents for the Three Studied Samples of 1, As Determined from Calorimetric Measurements sample

Theating (K)

ΔHheating (kJ/mol)

ΔSheating (J/mol K)

Tcooling (K)

ΔHcooling (kJ/mol)

ΔScooling (J/mol K)

Theating (K) 1/2

Tcooling (K) 1/2

PP APP SC

299.1 298.1 296.2/301.2

19.3 21.3 21

66 72 72

283.9 281.3 279.2

20.5 22.8 23

75 82 84

299.2 297.6 296

282.4 279.8 277.8

Figure 3. Thermal hysteresis loops for: (A) PP sample, (B) aged PP sample (APP), and (C) SC sample. Irradiation temperatures (Tirradiation) for each sample are marked in the respective hysteresis loop and correspond to 289.15 K (P1, P1*), 291.15 K (P2*), 293.15 K (P3*), and 284.15 K (SC1) for the case of LS to HS excitation, while excitation at 287.15 K (P2), 286.65 K (P3), 285.15 K (P4), 284.15 K (P5), 286.15 K (SC2), and 283.15 K (SC3) was done for the HS to LS one. For the APP sample, excitation was also performed at 289.15 K (P4* from heating branch, P5* from cooling branch) when starting from a mixed HS/LS state.

measurements are fully reproducible and depict the first-order spin transition exhibited by this compound. Temperatures corresponding to the maximum of the peaks together with their enthalpy and entropy contents are gathered in Table 1. The three samples show strong but rather broad anomalies on cooling, corresponding to the HS to LS transition; however, on heating, for the LS to HS transition, a sharper peak is detected for the PP sample, and the SC sample presents two sharp overlapped peaks. The peaks for the APP sample are broader and shifted to slightly lower temperatures. The hysteresis width of the SC sample is around 20 K, similar to that for the polycrystalline samples (15−17 K). Critical temperatures are in the low temperature range of the previously reported temperatures on 1. Magnetic measurements (see Supporting Information, Figures S2 and S3) performed on the SC sample are in agreement with the calorimetric data. Additionally, single-crystal X-ray diffraction on the SC sample confirmed the previously reported structure, with no significant trace of reabsorbed lattice water molecules (see Supporting Information, Table S1 and Figure S5), and Fe−N bond lengths characteristic of HS and LS state, respectively, at 310 and 240 K. For the three samples, the presence of significant amounts of reabsorbed water can be discarded since the spin transition occurs at higher temperature than the fully hydrated sample.7,16 Figures 3A, B, and C show the thermal hysteresis loops determined from our calorimetric measurements after subtracting the appropriate baselines for the PP, APP, and SC samples, respectively. The HS fraction, α(T), has been obtained as the ratio between the enthalpy content of the photoinduced anomaly, ΔH(T), calculated by integration of the excess heat flow as temperature increases and the enthalpy content of the complete transformation, that is ΔHheating for the LS → HS case and ΔHcooling for the HS → LS one. Critical temperatures,

Theating and Tcooling 1/2 1/2 , determined from these hysteresis loops for the three samples are collected in Table 1. Irradiation experiments were performed using 532 nm laser pulses of varying fluences at temperatures Tirradiation, depicted in Figures 3A, B, and C, chosen to have similar temperature differences with respect to the characteristic temperatures (ΔT = T1/2heating − Tirradiation or Tirradiation − T1/2cooling) as in the relevant reported works7−9 in order to allow comparison. Let us first describe, following the example for the PP sample shown in Figure 4A, the experimental procedure for the photoexcitation of the LS species in order to induce the HS state. After cooling from room temperature to a low enough temperature (253 K for PP and APP samples and 243 K for the SC sample) to reach a complete transformation to the LS state (a), the sample is heated to Tirradiation inside the hysteresis loop (b) (for example P1, P1*, and SC1 point in Figure 3A, B, and C, respectively, then the sample is still fully LS), and this temperature is maintained constant (c). During this isotherm, a single laser pulse (or several of them) is applied to the sample. Figure 4A shows the case of a single laser pulse applied at 8 min with a laser fluence of 27 mJ/cm2. A peak on the heat flow is clearly detected corresponding to the response of the calorimeter in order to keep a constant temperature against the effect of the laser pulse that disturbs this isothermal step (see rectangular box in Figure 4A). Finally, the thermogram is recorded on cooling (d), and an anomaly, centered at around Tcooling, is detected corresponding to the conversion of HS species to the LS state (see circle in Figure 4A and thick line in Figure 4B), thus providing direct evidence that HS species have been produced by the laser pulse. To determine the HS fraction produced by the irradiation, α, defined here as the ratio between the enthalpy content of the anomaly, ΔHPI, and the enthalpy content of the complete HS D

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Figure 4. (A) Example of the programmed temperature profile (thick line) and heat flow (thin line) during the experimental procedure, as explained in the text, in order to perform photoexcitation of the LS state. The dashed line represents heat flow equal to zero. Rectangular box: Peak on the heat flow corresponding to the response of the calorimeter in order to keep the sample temperature constant when the laser pulse is applied. Circle: Calorimetric anomaly corresponding to the conversion of the photoinduced HS species to LS state. (B) Process followed (see text) to deduce the enthalpy content of the photoinduced anomaly, ΔHPI, needed to estimate the HS fraction induced by the irradiation. The thick line is the calorimetric anomaly corresponding to the conversion of the photoinduced HS species to LS state, and the thin line is the background heat flow (BHF).

Figure 5. (A) Calorimetric anomalies for the PP sample after irradiation at 289.15 K (P1, 100% LS state). Excess heat flow (ΔHeat flow) after subtracting the corresponding background heat flow (BHF) for two laser fluencies (27 (I) and 60 (II) mJ/cm2) and also for the case of 4(III), 8(IV), and 12 (V) consecutive pulses of 60 mJ/cm2. (B) HS fraction induced by irradiation versus laser fluence on compound 1. Experimental results from the literature are also represented for comparison (see text). Arrows indicate the HS fraction increase when several laser pulses are applied consecutively.

spectroscopy are gathered on the same scale to allow proper comparison. For the PP sample, no excitation is detected for a laser pulse of around 2.5 mJ/cm2, but α reaches 0.13 and 0.33 for 27 and 60 mJ/cm 2 , respectively, with a good reproducibility. Consecutive pulses of 60 mJ/cm2 allow increasing the photoinduced fraction up to a saturation value of 0.60, for 12 successive pulses. Irradiation of the APP sample with pulses of 38, 50, and 60 mJ/cm2 resulted in slightly less efficient transformations than for the PP sample (ca. 30%, see Figure S6, Supporting Information), but these data, scaled to the PP values at 60 mJ/cm2, actually follow the exact same trend as shown in Figure 5B where scaled data are depicted. The results for the SC sample are comparable since for 38 and 60 mJ/cm2 α is 0.18 and 0.33, respectively (see Supporting Information, Figure S7). In order to check the consistency of our results, α has been additionally determined, for the latter case, by measuring a heating thermogram instead of a cooling one since the calorimetric anomaly due to the transition from LS to HS after the irradiation must be smaller than the case without light

→ LS transformation ΔHcooling (Table 1), a baseline must be subtracted. This background heat flow, BHF, indicated in Figure 4A and shown in Figure 4B (as a thin line) together with the anomaly for the case of a laser fluence of 27 mJ/cm2, is obtained subjecting the sample to the exact same thermal history (see Figure 4A: e−h) but without the application of the laser pulse. The excess heat flow (ΔHeat flow) after subtracting the corresponding baseline and for the different irradiation experiments performed at 289.15 K (100% LS state corresponding to point P1 in Figure 3A) is shown in Figure 5A for the PP sample together with their enthalpy contents. Similarly, excess heat flow for the case of the APP (for P1* in Figure 3B) and SC (for SC1 in Figure 3C) samples is shown in Figures S6 and S7 (Supporting Information), respectively. The results for the three samples are summarized in Figure 5B where α versus the laser fluence is represented and ΔT = 10, 8.5, and 12 K for the PP, APP, and SC samples, respectively. The available data from previous studies on 1 using Raman E

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

m*·ΔHheating gives m* = 0.35 mg, at most 32% of the total PP sample mass (m =1.10 mg) would be transformed with one laser pulse. At first the fact that this estimated transformed fraction is close to the observed one, 33%, seems to indicate a dominant role played by thermal effects. However, the calculation is made neglecting optical transformation and most of all without considering any heat transfer to the thermal bath nor to the rest of the sample. A detailed evaluation of these effects requires analyzing a complex thermal model, a task that is beyond this study. In our case, the heat transfer to the thermal bath is particularly relevant since the sample is in relatively good thermal contact with large masses of metallic parts (mainly the aluminum pan and also the sample support of the DSC oven), in addition maintained at constant temperature by the setup. Actually, due to the laser pulse, only a small perturbation of the measured temperature well below 0.5 K is detected by the thermocouple located in the sample support and in contact with the bottom of the aluminum pan. This means that a significant fraction of the energy deposited in the sample by the laser pulse will not be invested in increasing the sample temperature. The typically low thermal conductivity of coordination compounds32 also means that the propagation of the temperature increase within the sample will be slow and thus allow for heat losses to the environment. Eventually, any portion of the sample suffering a smaller ΔTh will either transform only partially (if reaching the warming branch of the hysteresis loop) or not at all. Altogether, it therefore appears evident that the observed transformation of 33% can by no means be solely due to a heating effect. This statement is also supported by the fact that similar transformed fractions are obtained when irradiating with a laser pulse of 60 mJ/cm2 the APP sample at different temperatures (P1*, P2*, and P3*, that is ΔT = 8.5, 6.5, and 4.5 K respectively), while an increase would be expected upon reducing ΔT (i.e., performing the excitation closer to Theating 1/2 ) if the heating-induced mechanism would be dominant. Additionally, the saturation effect obtained using several pulses can also be explained by absorption and not only by thermal effects.15 Due to the fact that the LS state has a higher absorption than the HS state for the excitation wavelength (see Figure S4, Supporting Information) and as the LS species at the particle/sample surface decrease, the penetration increases, and more HS species could be photoinduced until a saturation is reached when the intensity is reduced by the HS state absorption and no further excitation of LS can be induced. Therefore, while there is for sure some heating effect involved, it appears that the LS to HS conversion within the thermal hysteresis of [Fe(pyrazine)Pt(CN)4] is not the (sole) result of a local heating due to the laser pulses used, in line with previous affirmations.7−9 Altogether our results are in agreement with the original data of ref 7, but differ with the later reported results by the same authors (see Figure 5B) on the same compound 9 and with those on [Fe(PMBiA) 2 (NCS) 2 ] and [Fe(NH 2 -trz) 3 ](NO 3 ) 2 ·H 2 O compounds5,15 where an increase of the high spin fraction as the excitation temperature increases has been determined. These discrepancies are striking, although a number of experimental conditions, some difficult to control or compare, may have a non-negligible effect, such as the material composition and properties, sample characteristics (crystallinity, crystallite size), or the thermal contacts and temperature controls. We have also investigated the excitation inside the hysteresis loop from the HS state using the same type of irradiation. In this case, the sample was cooled from room temperature (HS

excitation (see Supporting Information, Figure S8). In this case, α amounts to 0.28, in agreement with the value obtained using the cooling procedure, 0.33. The photoinduced HS fractions obtained here are comparable to the first reported results with a polycrystalline sample7 where from the LS state and after irradiation with one single laser pulse with very large fluence of 250 mJ/cm2 (ΔT = 2.5 K) a HS fraction of 0.2 has been reached, 0.4 after a second pulse is applied. In our case, a laser fluence of only 40−60 mJ/ cm2 is required to reach a converted HS fraction of 0.2−0.3, and successive pulses improve the overall conversion to 0.60. On the contrary, in the original work, no further increment of the HS fraction was produced by additional pulses, but similar results were obtained with ΔT = 7, 15, and 22 K performing the excitation with 10 successive pulses.7 However, the same group later reported a complete photoconversion8,9 for both polycrystalline (ΔT = 8 K) and single-crystal (ΔT = 6 K) samples from the LS to the HS states using a single pulse and with only laser fluences above 20 mJ/cm2 and with a threshold value of 5 mJ/cm2. Our experiments performed at 2.5 mJ/cm2 agree with the presence of a threshold but do not reproduce this complete photoconversion. A possible explanation may lie in the larger thickness of our polycrystalline sample since the radiation may not reach the whole sample and therefore cannot completely excite it.18 Another source of discrepancy may come from the determination of the laser fluences, which we determined here in situ using the calorimeter (see Supporting Information), while it is not described in the related reports.7−9 Another aspect relevant to the efficiency of the transformation lies in possible heating effects associated with the laser pulse and the energy deposited in the sample, as well as the thermal control of the used setup during the photoirradiation experiments. In fact, a large transient temperature jump, ΔTh, may occur, at least locally, and some authors have estimated it to be of some tens of degrees, therefore allowing them to claim that the heating mechanism plays a key role in the LS to HS conversion.16,18,19,21 Even in the case of a bistable system, such a transient temperature incursion may be large enough to allow the sample to reach or surpass the ascending branch of the hysteresis. To assess this possibility in our case, let us roughly evaluate the maximal transformed fraction, i.e., considering the minimal necessary ΔTh, assuming the photoinduced effect is solely thermal and the whole pulse energy is invested to increase the temperature of the sample. Of course, this also has to involve the energy required to convert the LS species into HS ones and thus the total enthalpy associated with the transformation, ΔHheating (see Table 1), a term particularly important due to the first-order character of the spin transition. This contribution has not been considered by some authors16,21 for the estimation of the temperature jump for first-order spin transition, and only the heat capacity term, enough in the case of second-order transition,18,19,21 has been used. In our case, for a single pulse with a laser fluence of 60 mJ/cm2 applied at 289.15 K, the total amount of energy reaching the sample is E = 17 mJ, assuming that the sample is spread uniformly on the aluminum capsule (area = π·0.32 = 0.283 cm2). For the PP sample, a ΔTh of 11 K would be sufficient to result in a full LS to HS transformation and ΔHheating = 44.34 J/g (see Table 1 and using 435.3 g/mol as molecular mass). Then considering a mean value of the heat capacity for this compound31 in the 270−320 K temperature range (excluding the transition temperature range) as Cp = 0.378 J/g K and the simple relation E = m*·Cp·ΔTh + F

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

fact that the enthalpy content of the detected calorimetric anomalies (equal with and without laser pulse) is larger as the temperature of irradiation is lower supports the thermal origin for these anomalies and reflects the conversion of the preexisting LS species at the irradiation temperature to the HS state. The use of large energetic pulses (16, 27, and 60 mJ/cm2 at P3, P4, and SC2, respectively) was also tested but does not allow inducing a detectable amount of LS species. The present calorimetric study therefore evidences that no HS to LS optical conversion can be obtained as the result of nanosecond-pulsed green laser irradiation, in agreement with the absence of an absorption band in this range of wavelengths. The previously detected presence of LS species under similar conditions most likely arises from thermally converted centers due to the closeness of the experiment temperature to the thermal conversion. Here, it is interesting to note that Hauser et al. have recently produced an efficient HS to LS switching (reverse-LIESST) in the vicinity of the thermal transition of the [Fe(ptz)6](BF4)2 spin crossover compound (ptz = 1-propyltetrazole) using an adequate wavelength of 830 nm and unraveling the role of ligand-field states in the ultrafast photophysical cycle of this specific compound.33 In order to evaluate the influence of the initial state on the LS to HS excitation process, experiments were repeated starting from a thermally induced HS/LS mixed state instead of a full LS state. For this, the APP sample is first heated from the LS state up to 298.15 K, reaching a HS/LS mixed state with α = 0.6 according to the hysteresis loop (Figure 3B). After the sample temperature is decreased down to 289.15 K (P4* in Figure 3B), a pulse of 60 mJ/cm2 is applied, and a thermogram is performed on cooling, allowing the corresponding anomaly to be recorded. The experiment following the same thermal history, but without the application of the pulse, has also been performed for comparison. The results are shown in Figure 7A together with the peak corresponding to the complete HS → LS transition. Without pulse, an anomaly centered at 285 K is obtained with a low-temperature tail down to 270 K. The peak is significantly shifted to higher temperature than the peak corresponding to the complete HS → LS transition in agreement with the expected behavior when a minor loop is followed and thus consistent with the formation of independent domains.34 Additionally, considering the enthalpy content we can estimate a high spin fraction of 0.56 in agreement with the expected value at 298.15 K from the heating branch of the hysteresis loop. The excitation changes the profile of the peak, increasing the tail and decreasing the contribution at higher temperature. In this case, we can evaluate α = 0.69, and then an increase of 0.13 has been induced by the irradiation, which is lower than the value obtained by the same pulse performed on the 100% LS state and at the same temperature. The same experiment was repeated with four pulses, and again the profile of the peak was modified showing a larger tail at low temperature corresponding to a prominent peak and a decrease of the peak at high temperature with a maximum temperature shifted to lower temperature. The HS fraction increase was only 18%, smaller than the 50% achieved from the 100% LS state. We have then performed the same type of experiment but creating the HS/LS mixed state from the pure HS, i.e., by cooling the sample from the HS state down to 280.15 K (α = 0.55 from the cooling branch of the hysteresis loop in Figure 3B), then heating to 289.15 K (P5* in Figure 3B), irradiating with a laser pulse of 60 mJ/cm2, and recording a thermogram on cooling. The same thermal process was repeated but without

state) down to a temperature (P2−P5 in Figure 3A for the PP sample and SC2 and SC3 for the SC sample in Figure 3C) where the compound is still in the HS state or presents a HS− LS mixture state. Once the temperature is stable and controlled by the calorimeter, a laser pulse is applied to induce LS species. Subsequently, a heating thermogram is performed in order to detect a calorimetric anomaly due to the conversion of these LS species to a HS state. Because the energy of the pulse seems to be crucial in order to induce LS species, i.e., high energetic pulses were not suitable to photoinduce them from the HS state because heating of the sample competes against the photoexcitation,8 we first performed the experiments using a low laser fluence of around 2.5 mJ/cm2. As an example, Figure 6 shows one of these experiments performed in the PP sample.

Figure 6. Calorimetric experiments corresponding to the irradiation at 287.15 K (P2) on the PP sample using a low laser fluence (2.5 mJ/ cm2). Sample has been subjected to the same thermal history with and without irradiation. A similar anomaly is obtained in the subsequent heating, indicating that no photoinduced LS species are produced.

After the irradiation at P2 (see Figure 3A), only a very small anomaly is detected in the subsequent heating thermogram at around 300 K in agreement with the expected temperature range if some amount of LS species was induced and these were converted into HS species. However, when the same experiment is performed without the application of the laser pulse, the calorimetric anomaly also appears with comparable enthalpy content, indicating that the LS photoinduced fraction is negligible. A similar result has been obtained for the SC sample when irradiation was performed at SC2 (see Figure 3C and Supporting Information Figures S9 and S10). This behavior is in disagreement with ref 8 where a complete HS to LS switch was induced. On the other hand, our results do agree with those reported in ref 7 (our experimental conditions are similar to the experiment corresponding to point H of Figure 5 in this reference) where no excitation of LS species has been obtained from the HS state. The same behavior has been obtained for the irradiation at P3, P4, P5, and SC3 (see Supporting Information, Figures S11−S15) where the excitation is performed in a LS−HS mixture state. This result also disagrees with ref 8 since when the irradiated state is a mixture of LS and HS species and a low power laser pulse is used some amount of photoinduced LS species is obtained. The G

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

HS/LS mixed state reached through the heating branch but still significantly lower than the value reached from the full LS state corresponding to α = 0.26. However, when the excitation is performed at the same temperature on a similar HS/LS mixed state obtained previously by irradiation (P1 in Figure 3A and after four pulses a HS fraction of 0.5 is reached from the 100% LS state close to the previous thermally induced mixed HS/LS state) the increase of the HS fraction by four additional consecutive pulses was only 0.08. It thus appears that irradiation from a thermally induced mixed HS/LS state from either HS and LS state is less efficient than from a pure LS state but more efficient than from a similar mixed HS/LS state reached by irradiation.



CONCLUSIONS



ASSOCIATED CONTENT

In summary, the optical switching from LS to HS state of the [Fe(pz)Pt(CN)4] SCO material through single and multiple green (λ = 532 nm) nanosecond-laser pulse(s) has been confirmed and accurately evaluated using calorimetry, a complement to the Raman spectroscopy-only results previously reported. A maximum conversion of 60% has been obtained by successive pulses, likely limited by the penetration depth of light within the sample/crystallites. Besides, the results seem to indicate that, although thermal effects may be present, the LS to HS transformation is not the sole consequence of such effects. On the other hand, the reverse transformation, HS to LS, previously reported to be induced at the same wavelength, has not been observed, and evidence is given here that such a transformation is likely the result of thermal effects and/or relaxation of the HS species, due to the closeness of the temperature experiments to the cooling branch of the thermal SCO hysteresis. Finally, conversion starting from a thermally induced LS/HS mixed state is found to be less efficient than from a pure LS state, but more efficient than from a similar mixed LS/HS state obtained by irradiation. This work highlights the great potential of photocalorimetry to study switching of cooperative SCO materials within their bistability domain.

Figure 7. Anomalies (crosses) obtained on cooling after excitation with a laser pulse of 60 mJ/cm2 at 289.15 K from a HS/LS mixed state reached, as explained in the text, from the 100% LS (A) and 100% HS (B) states. The complete HS to LS transition (small dots) and the anomalies obtained following the same thermal history but without laser pulse (open circles) are also shown for comparison (see text).

S Supporting Information *

Determination of the laser fluences. Temperature dependence of the magnetic properties of the SC sample of 1. Visible absorption spectra of powder sample for the LS and HS state. Crystallographic data and refinement details of 1 at 310 and 240 K. Selected calorimetric experiments with irradiation. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.5b05864.

the application of the pulse in order to have a reference baseline. The latter, shown in Figure 7B, presents an anomaly starting at around 280 K as expected and matches perfectly the low temperature part of the peak corresponding to the complete HS to LS transition. This anomaly corresponds to the final conversion of the remaining HS species. Considering its enthalpy content (Figure 7B), the converted HS fraction was 0.35, lower than the expected value of 0.55. This can be due to a relaxation process of the HS centers to the ground LS state, similar to the case of the [Fe(H4L)2](ClO4)2·H2O·2(CH3)2CO SCO compound.10 Irradiation modifies this anomaly, shown also in Figure 7B, and a new peak appears at the high temperature side and also reproduces the peak obtained without the application of the pulse. Both anomalies are perfectly delimited by the peak corresponding to the 100% HS to LS transition. The mixed HS/LS state reached corresponds to α = 0.52, and then the increase due to the irradiation is 0.17 which is slightly higher than the HS fraction increase from the



AUTHOR INFORMATION

Corresponding Author

*Tel.: 00-34-976762528. Fax: 00-34-976761957. E-mail: [email protected]. Author Contributions

LPL and JAR synthesized the materials. JARV, OR, and MC realized the experiments. The manuscript was written by OR and MC with contributions from all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest. H

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C



Crossover Compound inside Its Thermal Hysteresis Loop. Appl. Phys. Lett. 2010, 96, 041907. (15) Gallé, G.; Degert, J.; Mauriac, C.; Etrillard, C.; Létard, J.-F.; Freysz, E. Nanosecond Study of Spin State Transition Induced by a Single Nanosecond Laser Shot on [Fe(NH2trz)3] Compounds inside and outside Their Thermal Hysteresis Loops. Chem. Phys. Lett. 2010, 500, 18−22. (16) Fouché, O.; Degert, J.; Jonusauskas, G.; Daro, N.; Létard, J.-F.; Freysz, E. Mechanism for Optical Switching of the Spin Crossover [Fe(NH2-trz)3](Br)2·3H2O Compound at Room Temperature. Phys. Chem. Chem. Phys. 2010, 12, 3044−3052. (17) Guillaume, F.; Tobon, Y. A.; Bonhommeau, S.; Létard, J.-F.; Moulet, L.; Freysz, E. Photoswitching of the Spin Crossover Polymeric Material [Fe (Htrz)2(trz)] (BF4) under Continuous Laser Irradiation in a Raman Scattering Experiment. Chem. Phys. Lett. 2014, 604, 105− 109. (18) Lorenc, M.; Hébert, J.; Moisan, N.; Trzop, E.; Servol, M.; Buron-Le Cointe, M.; Cailleau, H.; Boillot, M. L.; Pontecorvo, E.; Wulff, M.; et al. Successive Dynamical Steps of Photoinduced Switching of a Molecular Fe(III) Spin-Crossover Material by TimeResolved X-Ray Diffraction. Phys. Rev. Lett. 2009, 103, 028301 . Note that a finite penetration depth of the order of tens of microns has been considered in this work. . (19) Lorenc, M.; Balde, Ch.; Kaszub, W.; Tissot, A.; Moisan, N.; Servol, M.; Buron-Le Cointe, M.; Cailleau, H.; Chasle, P.; Czarnecki, P.; et al. Cascading Photo-induced, Elastic and Thermal Switching of Spin States Trigerred by a Femtosecond Laser Pulse in an Fe(III) Molecular Crystal. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85, 054302. (20) Tissot, A.; Bertoni, R.; Collet, E.; Toupet, L.; Boillot, M.-L. The Cooperative Spin-State Transition of a Iron (III) Compound [FeIII (3MeO-SalEen)2]PF6: Thermal- vs. Ultra-Fast Photo-Switching. J. Mater. Chem. 2011, 21, 18347. (21) Bertoni, R.; Lorenc, M.; Tissot, A.; Boillot, M.-L.; Collet, E. Femtosecond Photo-Switching Dynamics and Microsecond Thermal Conversion Driven by Laser Heating in FeIII Spin-Crossover Solids. Coord. Chem. Rev. 2015, 282−283, 66−76. (22) Bedoui, S.; Lopes, M.; Nicolazzi, W.; Bonnet, S.; Zheng, S.; Molnár, G.; Bousseksou, A. Triggering a Phase Transition by a Spatially Localized Laser Pulse: Role of the Strain. Phys. Rev. Lett. 2012, 109, 135702. (23) Niel, V.; Martinez-Agudo, J. M.; Muñoz, M. C.; Gaspar, A. B.; Real, J. A. Cooperative Spin Crossover Behavior in Cyanide-Bridged Fe(II)-M(II) Bimetallic 3D Hofmann-Like Networks (M = Ni, Pd and Pt). Inorg. Chem. 2001, 40, 3838−3839. (24) CrysAlis Pro; Agilent Technologies XRD Products: Yarnton, U.K., 2012. (25) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. SIR97: A New Tool for Crystal Structure Determination and Refinement. J. Appl. Crystallogr. 1999, 32, 115−119. (26) Sheldrick, G. M. Crystal Structure Refinement with SHELXL. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8. (27) Sheldrick, G. M. SHELXTL; Bruker AXS Inc.: Madison, WI, USA, 2014. (28) Tayagaki, T.; Galet, A.; Molnár, G.; Muñoz, M. C.; Zwick, A.; Tanaka, K.; Real, J. A.; Bousseksou, A. Metal Dilution Effects on the Spin-Crossover Properties of the Three-Dimensional Coordination Polymer Fe(pyrazine)[Pt(CN)4]. J. Phys. Chem. B 2005, 109, 14859− 14867. (29) (a) Ohba, M.; Yoneda, K.; Agustí, G.; Muñoz, M. C.; Gaspar, A. B.; Real, J. A.; Yamasaki, M.; Ando, H.; Nakao, Y.; Sakaki, S.; et al. Bidirectional Chemo-Switching of Spin State in a Microporous Framework. Angew. Chem., Int. Ed. 2009, 48, 4767−4771. (b) Agustí, G.; Ohtani, R.; Yoneda, K.; Gaspar, A. B.; Ohba, M.; Sánchez-Royo, J. F.; Muñoz, J. F.; Kitagawa, S.; Real, J. A. Oxidative Addition of Halogens on Open Metal Sites in a Microporous SpinCrossover Coordination Polymer. Angew. Chem., Int. Ed. 2009, 48, 8944−8947.

ACKNOWLEDGMENTS This work has been supported by the Spanish MINECO and European Commission FEDER under projects MAT201124284, MAT2011-27233-C02-02, CTQ2011-23862-C02-01, CSD2007-00010, and CTQ2013-46275-P, by the Gobierno de Aragón through grants E98-“MOLCHIP” and E100“Propiedades Térmicas de Materiales”, and by the Generalitat Valenciana through PROMETEO/2012/049. L.P.L. thanks the Generalitat Valenciana for a predoctoral fellowship in the frame of the project PROMETEO/2012/049. JARV acknowledges CSIC for a JAE-doc contract. The authors acknowledge the use of the Physical measurements and X-ray diffraction units of Servicio General de Apoyo a la Investigación-SAI, Universidad de Zaragoza, and of the differencial scanning calorimeter Q1000 of Servicio de análisis térmico y calorimetriá del Instituto de Ciencia de Materiales de Aragón (ICMA).



REFERENCES

(1) Sato, O.; Tao, J.; Zhang, Y. Z. Control of Magnetic Properties through External Stimuli. Angew. Chem., Int. Ed. 2007, 46, 2152−2187. (2) Gütlich, P.; Goodwin, H. A. Topics in Current Chemistry. Spin Crossover in Transition Metal Compounds; Springer-Verlag: Berlin, 2004; Vol. 233−235. (3) Bousseksou, A.; Molnár, G.; Salmon, L.; Nicolazzi, W. Molecular Spin Crossover Phenomenon: Recent Achievements and Prospects. Chem. Soc. Rev. 2011, 40, 3313−3335. (4) Hauser, A. Light-Induced Spin Crossover and the High-Spin Low-Spin Relaxation. Top. Curr. Chem. 2004, 234, 155−198. (5) Freysz, E.; Montant, S.; Létard, S.; Létard, J.-F. Single Laser Pulse Induces Spin State Transition within the Hysteresis Loop of an Iron Compound. Chem. Phys. Lett. 2004, 394, 318−323. (6) Liu, H.; Fujishima, A.; Sato, O. Photoinduced Phase Transition in an Iron (II) Spin-Crossover Complex with a N3O2 Macrocyclic Ligand. Appl. Phys. Lett. 2005, 86, 122511. (7) Bonhommeau, S.; Molnár, G.; Galet, A.; Zwick, A.; Real, J. A.; McGarvey, J. J.; Bousseksou, A. One Shot Laser Pulse Induced Reversible Spin Transition in the Spin-Crossover Complex [Fe(C4H4N2){Pt(CN)4}] at Room Temperature. Angew. Chem., Int. Ed. 2005, 44, 4069−4073. (8) Cobo, S.; Ostrovskii, D.; Bonhommeau, S.; Vendier, L.; Molnár, G.; Salmon, L.; Tanaka, K.; Bousseksou, A. Single-Laser-Shot-Induced Complete Bidirectional Spin Transition at Room Temperature in Single Crystals of (FeII(pyrazine)[Pt(CN)4]. J. Am. Chem. Soc. 2008, 130, 9019−9024. (9) Bonhommeau, S.; Molnár, G.; Cobo, S.; Ostrovskii, D.; Bousseksou, A. One Laser Shot Induces Complete Phase Transition in the Spin Crossover Complex Fe(pyrazine)[Pt(CN)4]. Polyhedron 2009, 28, 1610−1613. (10) Craig, G. A.; Sánchez Costa, J.; Roubeau, O.; Teat, S. J.; Shepherd, H. J.; Lopes, M.; Molnár, G.; Bousseksou, A.; Aromí, G. High-Temperature Photo-Induced Switching and Pressure-Induced Transition in a Cooperative Molecular Spin-Crossover Material. Dalton Transactions 2014, 43, 729−737. (11) Roubeau, O. Triazole-Based One-Dimensional Spin-Crossover Coordination Polymers. Chem. - Eur. J. 2012, 18, 15230−15244. (12) Moussa, N. O.; Ostrovskii, D.; Martinez Garcia, V.; Molnár, G.; Tanaka, K.; Gaspar, A. B.; Real, J. A.; Bousseksou, A. Bidireccional Photo-Switching of the Spin State of Iron (II) Ions in a Triazol Based Spin Crossover Complex within the Thermal Hysteresis Loop. Chem. Phys. Lett. 2009, 477, 156−159. (13) Fouché, O.; Degert, J.; Jonusauskas, G.; Baldé, C.; Desplanches, C.; Létard, J.-F.; Freysz, E. Laser Induced Spin State Transition: Spectral and Temporal Evolution. Chem. Phys. Lett. 2009, 469, 274− 278. (14) Gallé, G.; Deldicque, D.; Degert, J.; Forestier, T.; Létard, J.-F.; Freysz, E. Room Temperature Study of the Optical Switching of a Spin I

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C (30) Rodríguez-Velamazán, J. A.; González, M. A.; Real, J. A.; Castro, M.; Muñoz, M. C.; Gaspar, A. B.; Ohtani, R.; Ohba, M.; Yoneda, K.; Hijikata, Y.; et al. A switchable Molecular Rotator: Neutron Spectroscopy Study on a Polymeric Spin-Crossover Compound. J. Am. Chem. Soc. 2012, 134, 5083−5089. (31) Palacios E., Private Communication. Heat Capacity of the [Fe(pyrazine)(Pt(CN)4)] Compound Measured by Adiabatic Calorimetry. 2015. (32) Jakobi, R.; Romstedt, H.; Spiering, H.; Gütlich, P. Microcalorimetry of the Spin-Transition Systems [Fe(bts)2(NCS)2] and [Fe(2-pic)3Cl2]·EtOH in the Range 15 K < T < 300 K. Angew. Chem., Int. Ed. Engl. 1992, 31, 178−180. (33) Marino, A.; Chakraborty, P.; Servol, M.; Lorenc, M.; Collet, E.; Hauser, A. The Role of Ligand-Field States in the Ultrafast Photophysical Cycle of the Prototypical Iron (II) Spin-Crossover Compound [Fe(ptz)6](BF4)2. Angew. Chem., Int. Ed. 2014, 53, 3863− 3867. (34) (a) König, E.; Ritter, G.; Kulshreshtha, S. K. The Nature of Spin-State Transition in Solid Complexes of Iron (II) and the Interpretation of Some Associated Phenomena. Chem. Rev. 1985, 85, 219−234. (b) Koenig, E.; Kanellakopulos, B.; Powietzka, B.; Goodwin, H. A. Detailed Study of the Hysteresis Associated with the Spin-State Transition in Bis(2,6-bis(pyrazol-3-yl) pyridine) Iron (II) bis(tetrafluoroborate) and the Hypothesis of Independent Domains. Inorg. Chem. 1989, 28, 3993−3996. (c) Koenig, E.; Kanellakopulos, B.; Powietzka, B.; Goodwin, H. A. Detailed Hysteresis Studies and Nature of the Spin-State Transition in bis(1,10-phenanthroline-2-carbaldehyde phenylhydrazone) Iron (II) Complexes. Inorg. Chem. 1990, 29, 4944−4949. (d) Constant-Machado, H.; Stancu, A.; Linarès, J.; Varret, F. Thermal Hysteresis Loops in Spin-Crossover Compounds Analyzed in terms of Classical Preisach Model. IEEE Trans. Magn. 1998, 34, 2213−2219.

J

DOI: 10.1021/acs.jpcc.5b05864 J. Phys. Chem. C XXXX, XXX, XXX−XXX